555
Say you are taking in an audio input from a headphone speaker and you would like the audio to gradually decrease until inaudible.
Are there better, also very common chips that can do this more reliably?
Let's say you wanted the volume to decrease over 1 minute?
I notice from your circuit that the 555 and motors are running from a common (5V) power supply without any form of decoupling. The 555 is a current hog and draws a surge of current from the supply every time its output changes.
Try putting a large electrolytic (say 220uF) and a 0.1uf (helps with high frequencies) across each the Vcc and ground pin of the three 555s.
I would also make the capacitor at pin 5 a bit bigger (say 0.1uF) - its job is to smooth out any noise of the internal divider chain of 5K resistors. 47pF is way too low.
How can I debug ... I see no voltage on pins 2,6 and 7 to ground.
If I understand correctly, those are all fed via R1 so I'd look closely for a dry joint at R1. Even if C1 was shorted, I'd expect twiddling R2 would produce something measurable at pin 7.
If you disconnect power and pull the 555 out of the socket, I think you should probably be able to test c1,c2,r1,r2 in-circuit with a multimeter.
For example I'd expect resistance between socket positions 4 and 7 should be 1K
etc.
Best Answer
Honestly, one of the least fussy solutions would be a small MCU and a digipot. Specifically, a log digital potentiometer designed for audio (so that it switches on zero crossings), and just about any microcontroller. Most digital pots are controlled via I2C (or occasionally SPI).
You can trigger the "fade" however you like (button, etc), and you have very good control of fade length and linearity. You can even do stereo with nearly gain-matched channels (there are digipots specifically for stereo audio).
This way, no analog weirdness, programmable fade characteristics, flexible triggering (you still haven't said how you want it to trigger), and basically two parts (plus a power source and a smidgen of passives).